© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW Community BlogPrivacy & TermsContact
    1. Home
    3. Under Peer Review · Sparse Autoencoders for Pythia-70M-Deduped
    5. Pythia-70M-Deduped
    7. Residuals
    9. 1-RES-SM
    11. 24062
    Prev
    Next
    INDEX
    Explanations

    legal terminology related to court cases

    oai_token-act-pair · gpt-4o-mini
    Triggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    ctigges/pythia-70m-deduped__res-sm_processed/1-res-sm
    Prompts (Dashboard)
    32,768 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    32,768
    Data Type
    torch.float32
    Hook Name
    blocks.1.hook_resid_post
    Hook Layer
    1
    Architecture
    standard
    Context Size
    128
    Dataset
    EleutherAI/the_pile_deduplicated
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Neuron Alignment
    Index
    Value
    % of L₁
    55
    +0.14
    0.8%
    105
    +0.13
    0.8%
    270
    +0.13
    0.8%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    157
    +0.14
    0.03
    18
    +0.13
    0.02
    315
    +0.13
    0.02
    Negative Logits
    ?).
    -1.63
    cibly
    -1.59
    CDATA
    -1.56
     Engineers
    -1.44
    Figure
    -1.43
    fig
    -1.41
    ffected
    -1.40
    gow
    -1.40
    comings
    -1.37
    Fig
    -1.36
    POSITIVE LOGITS
    ®
    1.67
    igan
    1.59
     honor
    1.57
     headline
    1.52
    Ģ
    1.49
    athe
    1.48
     pounds
    1.42
    ±
    1.41
     editorial
    1.39
     conference
    1.39
    Activations Density 0.147%

    No Known Activations